The eligibility criteria involved identifying original empirical research published between 2016 and 2023 about blockchain and land administration or registration or real estate. The year 2016 was selected as the starting point because of the increasing momentum and a surge in interest on the potential applications of blockchain technology in land administration, with several pilot projects and proof-of-concept studies being conducted globally. The search consisted of the work of full-length peer-reviewed journals and conference papers published in English. As blockchain has gained attention in businesses and outside academia, grey literature was also included in the search. Within the context of this study, grey literature entailed blogs and commentary on legislative changes written by legal practitioners actively engaged in land registration processes, whose insights were considered valuable. Studies of any research design and methodology within any geographical context were also eligible for inclusion. Studies that focused on blockchain technology and land administration or registration in a fragmented manner were excluded and so were the viewpoints and commentaries without empirical findings. Studies of other disciplines such as cryptocurrency, Internet of Things (IoT), medicine and energy were excluded as were those published in languages other than English. Figure 1 demonstrates the inclusion and exclusion criteria used to inform the literature search. Lecture notes were also excluded from the search.

The search strategy for finding peer-reviewed journals and conference papers included the selection of the databases, namely, Mendeley, Scopus, ProQuest, Science Direct, Google Scholar and Research Gate, because of their accessibility via the University of South Africa’s (UNISA) library, which provided extensive coverage of peer-reviewed journals. These databases were selected because of their ability to provide technical insights into blockchain-based land registration. This assertion is reinforced by the fact that other researchers in the fields of technology and land registration also used similar databases for literature search (Ansah et al. 2023; Saari et al. 2022). Internet sources, presentation slides and previous systematic reviews were used to guide an electronic database search strategy. To refine the search and exclude irrelevant remarks, the search strategy used Boolean operators such as ‘AND’, ‘OR’ and ‘NOT’ along with the quotation marks in some of the databases. As such, the title/abstract/keyword (T/A/K) containing search terms such as ‘land registration’ OR ‘land administration’ OR ‘real estate’ OR ‘property registration’ NOT ‘land management’ AND ‘blockchain’ OR ‘block-chain’ OR ‘block chain’ OR ‘smart contracts’ were used to retrieve academic articles across different databases. Three rounds of searches were run from 22–24 March 2023, 6–17 November 2023 and 3–4 December 2023 to capture the relevant studies recently and since the initial search. The basic grouping of words consisted of blockchain technology OR smart contracts AND land registration OR real estate AND blockchain-based land registration.

In this study, the SLR followed a well-established four-step process flow included in Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). Full-text articles were retrieved from relevant databases and exported in Research Information Systems (RIS) format to Zotero, a reference management software and then exported to the library. A new review was created on Rayyan AI, a research collaboration platform for SLR, involving a single reviewer because of resource constraints. The review included a total of 258 references, which were sourced from various databases. To be specific, 178 were obtained from Mendeley, 44 were from Scopus, 25 were from Science Direct and 11 were from ProQuest, and they were successfully uploaded onto the software. During data review, 123 duplicates were detected, and subsequently, 24 references were resolved and 99 were deleted. The screening and reviewing process continued with 159 records, out of which 99 were deemed irrelevant for reasons such as incorrect publication type and unsuitable background article, leaving 60 reports that sought retrieval. Discrepancies were discussed between reviewers, and no disagreements occurred. Among these, 22 reports were inaccessible in PDF format, resulting in their exclusion. Eventually, 38 articles were assessed for eligibility, all of which met the criteria and were consequently included in the review, as illustrated in Figure 2. During the review and beyond the initial database search, backward and forward citation tracking was used to explore older and the most recent articles related to the topic, and this approach resulted in 14 additional studies added for analysis. Appendix 1 Table 1-A1 consists of a list of articles that were included in the study for review.

A high-quality review is complete and focuses on concepts. A complete review covers relevant literature on the topic and is not confined to one research methodology, one set of journals or one geographic region (Webster & Watson 2002). To guarantee and provide a high-quality review, this research describes a process for identification and inclusion of articles in a vigorous and detailed manner with the aim of encouraging an interested reader to replicate the procedure (Palmatier, Houston & Hulland 2018). Furthermore, the review follows the PRISMA 2020 guidelines outlined in the checklist to ensure the key aspects of the review are well documented to enhance transparency and replicability.

Thematic content analysis was conducted using the ATLAS.ti qualitative analysis software version 23.3 to facilitate the process of data coding. This step of the review began with data preparation and organisation, which included creating a new project on ATLAS.ti and uploading 52 documents from the desktop onto the software. Details pertaining to each article, including the topic, author and abstract, were included in the comment section of each document for ease of reference. Articles were further grouped and organised by the research methods chosen by the authors. This was followed by data familiarisation and immersion, which consisted of repeated reading and the analysis of articles, segmenting data into quotations by highlighting interesting passages of texts related to the research questions. Subsequently, preliminary codes were developed and assigned to the most frequently occurring words and phrases. Memos were simultaneously created to note the author’s thoughts and ideas generated from the articles, and these were linked to quotations and codes. The coded segments were further reviewed to identify preliminary themes from the emerged data based on recurring concepts appearing across multiple coded segments. Furthermore, code document analysis was performed to show the frequency of codes across document groups based on word count, as illustrated in Table 1.

Preliminary themes were reviewed and refined using data visualisation such as word clouds, word list and concepts for conceptual coding. This iterative approach led to the identification of common themes, discarding of irrelevant ones and merging of similar themes for a more streamlined and coherent analysis. Figure 3 is a word cloud illustrating the most commonly occurring words as ‘blockchain’, ‘technology’, ‘land’, ‘property’, ‘real estate’, ‘system’, ‘transactions’, ‘data’, ‘administration’ ‘property’, ‘information’, ‘registration’, ‘process’ and ‘smart contracts’ with the highest frequency. The word cloud was used to provide a snapshot of the commonly used phrases, which highlight the key topics and concepts that recur throughout the study and reflect the core themes and focus areas of the reviewed literature.

Table 2 presents the primary themes along with their associated coded segments identified from the code document analysis, while highlighting the emerging trends and patterns that inform the discussion of the results that are coming further (cf.) down. Five key themes emerged, namely, ‘System Organisation’, ‘Blockchain Core Features’, ‘Stakeholder Coordination’, ‘Process Effectiveness’ and ‘System Goals’. The themes and the codes identified are presented further below as a conceptual framework to interpret data relating to the economic benefits of the blockchain-based land registration system. This framework draws on findings from a range of studies analysing both pilot projects and full-scale implementations of blockchain in land registration processes.

Ethical clearance to conduct this study was obtained from the University of South Africa’s College of Economic and Management Sciences (CRERC Ref. No.4442).

To address the RQ 1 ‘how can GST provide a framework for understanding the application of blockchain technology in land registration as an integrated system?’, the review examines two themes identified in the literature: system organisation and stakeholder organisation. Mele, Pels and Polese (2010) addressed organisations as socio-technical systems, underlining the two main components of the firm as seen as a system: a social component (people) and a technical component (technology and machines). From this perspective, a land registration system is a social component of socio-technical systems because of the social relationship between members of the community with respect to land, which is an important element of consideration in any land administration (Ameyaw & De Vries 2023). On the contrary, the technical component can be attributed to blockchain technology because of its technical features necessary to solve the double-spending problem in the absence of trusted intermediaries (Ooi, Peng & Soh 2022). This is because blockchain provides a shared ledger technology that allows any participant in the network to see the one system of records (Panda et al. 2021). This is particularly applicable in the context of a blockchain-based land registration system, where the technical component represented by blockchain can provide data security and trust for the land registration system (Ansah 2023).

The organisation is seen as a system built by energetic input–output where the energy coming from the output reactivates the system (Mele et al. 2010). This concept of organisation is exemplified through the interconnection and interrelation of blocks via a hash function (Ali et al. 2020; Subedha et al. 2023; Tahar, Mendy & Ouya 2022). This process ensures that each block contains a reference to the preceding block, creating a secure and permanent record of transactions among multiple parties (Ali et al. 2020; Mendi 2020; Yadav & Kushwaba 2021). In this perspective, blockchain is considered as a system that can actively monitor the state of land, even when it is operationally inaccessible, by verifying the consistency of its own operations through a binary scheme which can record agreement or non-agreement (Luhmann 1986).

For competition and coordination to be combined, and for order to emerge, as well as for effective, efficient and equitable systems of human activity to evolve, a new organisational concept that entails the whole not controlling the parts, and none of the parts controlling the whole, is required (Hock is cited in Laszlo & Krippner 1998). This concept aligns with the decentralised nature of blockchain technology where the need for multiple intermediaries is eliminated, meaning no single party has complete control, allowing for a reduction in costs and increased transparency as all transactions can be made publicly available (Florentino & Bartolucci 2021). This is essentially important for socio-economic activities in the digital age, in which the right to data can be directly translated into business advantages (Du et al. 2023). A practical example of system organisation can be observed in the case of Georgia. Shang and Price (2019) argued that the success of the Georgia project stemmed from educating stakeholders about blockchain technology and the challenges it can address, including data quality. However, while blockchain technology possesses inherent features to enhance trust and transparency, the initial data entry still depends on human actors outside the system, and thus, blockchain cannot prevent fraud and incorrect data on first entry if the officials are corrupt (Lemieux 2017 is cited in Ansah et al. 2023).

This section of the review discusses the theme ‘blockchain core features’ to address RQ 2: ‘how can various aspects of the blockchain-based land registration system influence the country’s economic indicators’? Ooi et al. (2022) described the core features of blockchain as those necessary to solve the double-spending problem in the absence of intermediaries. Data security and trust are at the core of every data management system, and blockchain’s various characteristics play a crucial role in achieving this goal (Makala & Anand 2018; Saari et al. 2022).

The immutability feature of blockchain is observed in various countries by several authors. For example, Singh (2020) illustrated the landscape in India where nearly 66% of the court cases are related to land disputes, resulting in substantial litigation costs amounting to Rs.58 000 crore encompassing both civil and criminal proceedings. This is because the existing land records available in the country are not clear, poorly administered and, often, do not reflect the ground reality, and maintenance and availability of information are critical challenges faced by the government (Thakur et al. 2020). Consequently, India sought to ensure immutability of land data by employing the SHA256 cryptographic hash function and digital signatures to secure transactions between buyers and sellers (Zein & Twinomurinzi 2023). Furthermore, the digitisation of land registration can help boost the country’s GDP by as much as 1.3% (Singh 2020).

However, in Sweden, security provisions built into legacy systems have prevented further digitisation of the land transfer process, for example, by imposing restrictions through firewalls and limits on network connections. Consequently, while some institutional actors (e.g. realtors, banks and government agencies) could connect their systems to databases, individual buyers and sellers were disadvantaged and could not (McMurren, Young & Verhuist 2018).

Ghana, as highlighted by Mintah et al. (2020), experienced problems with skin land acquisition and title registration that emanated from the issuance of allocation notes, payment of kola money and use of physical ledger to document land transactions. The authors proposed a framework that would be a public register, arguing that information on all transactions on a specific parcel of land could be available to the public in real time, thus enhancing transparency and resolving the issue of encroachments and indeterminate land boundaries. Hence, stakeholders can determine rightful owners of land parcels before initiating transactions (Mintah et al. 2020).

Using the theme ‘process effectiveness’, the review seeks to address RQ 3: ‘what economic benefits of a blockchain-based land registration system, as identified from pilot programmes, can be anticipated for South Africa?’. Process effectiveness can be observed in the case of the Republic of Georgia where the transfer of land ownership suffered from significant challenges, including low efficiency and criminality in the form of corruption and bribery (Adil 2022). The adoption of blockchain in Georgia’s land registration resulted in process efficiency as it takes only one day to register land with registry cost amounting to a mere 0.1% of the property value (Shang & Price 2019). Blockchain has smoothed the problematic, time-consuming processes that were at risk of failure; it has made humans more powerful towards the implementation of transparency and accountability and in maintaining trust and security (Khalid et al. 2022).

Similarly, the United Arab of Emirates (UAE) faced challenges in traditional land registry management, leading to inefficiencies and delays in property transactions (Zein & Twinomurinzi 2023). Dubai’s strategic goal was to record all government transactions on blockchain, anticipating an estimated EUR1.2 billion saving per annum (Themistocleous 2018). Additionally, in 2016, a blockchain solution that records property history from end to end was designed to streamline processes through smart contracts and provide real-time traceability for each transaction (Zein & Twinomurinzi 2023).

In Sweden, the volume of documentation required in land transfer processes and the amount of duplication of data entry increase the risk of error, and between 4% and 7% of applications need to be resubmitted because of errors (McMurren et al. 2018). The intent for implementing blockchain into land registration was to reduce the time between the signing of a contract to purchase and the registration of the property title from 4 months to a few days; this was to be achieved through the elimination of steps in the process and by reducing delays brought about by mail and the need for repeated checks and physical signatures (McMurren et al. 2018). Rodima-Taylor (2021:144) reported that in 2018, the Ministry of Lands and Natural Resources in Ghana announced a partnership with the American technology and consulting company IBM to explore blockchain technology in the land sector and in other public institutions, for improved transparency and efficient service delivery.

Blockchain has the potential to help governments maintain their registers in an immutable and tamper-proof manner and increase the resilience of such registers generally ensuring the integrity of government records and services leading to a more powerful e-Government ecosystem, with reduced fraud, corruption, errors, possible cyber-attacks and paper-intensive administrative processes (Sladic et al. 2021). Literature reports developing countries as being dominated by fraud and corruption in their land registries (Saari et al. 2022). For example, the situation of Honduras where lack of a comprehensive country-wide land registry with valid and complete records, as well as political resistance to changing the status quo, put a stop to the blockchain project (Benbunan-Fich & Castellanos 2018).

Given that corruption often stems from manipulation, ambiguity and uncertainty of information processed, the decentralised, immutable and transparent features of blockchain technology make it particularly valuable for mitigating corrupt behaviour in different contexts (Gillpatrick, Boga & Aldanmaz 2022). Blockchain is unusual in that it is a social technology, designed to govern the behaviour of groups of people through social and financial incentives and therefore inherently political in a way that few other technologies are (Graglia & Mellon 2018). Consequently, author(s) in press proposes that a holistic view be adopted where various inputs (such as information, processes, finance and systems) from all stakeholders (including banks, conveyancers, central bank, deeds office, etc.) converge to produce a single output, which is property registration. As a result, as the author adds, cross-functional efficiencies and effectiveness will be realised.

On the contrary, developed countries such as the United States and the United Kingdom view blockchain for title recording system as the future for the maintenance of title records that can offer a solution in terms of the current recording system (Spielman 2016 is cited in Mata et al. 2022).

In conclusion, the conceptual framework provides a holistic overview of a blockchain-based land registration system as a socio-technical system. Central to the framework is a strategic use of blockchain key features and their benefits that have the potential to enhance process effectiveness. The framework suggests that the country should prioritise the economic benefits of a blockchain-based land registration as its primary system goals. Additionally, the framework highlights the crucial role of ongoing stakeholders’ coordination and engagement within a continuous feedback cycle to ensure effective implementation of the system.

Blockchain technology must overcome several challenges to be considered as reliable, legal and secure as the current real estate conveyancing system (Garcia-Teruel 2019). Literature has continued to highlight challenges in blockchain adoption, including implementation complexities, legal uncertainties and technical issues (Saari et al. 2022:14). Zein and Twinomurinzi (2023) suggested that addressing these socio-technical challenges is necessary for the successful implementation of blockchain, particularly in low-income countries. In the context of South Africa, attempts to implement technological innovation, modern systems and updates to operations to improve efficiency, quality and accuracy have encountered many challenges and costly delays (Williams-Wynn 2021).

The South African government has taken an initiative towards the electronic deed registration with the enactment of the e-DRSA that seeks to regulate the functions and laws related to title deed transfer, as well as the successful registration of title deeds. Consequently, the implementation of a blockchain-based land registration in South Africa necessitates careful consideration of several crucial factors despite its undeniable benefits such as immutability, high level of security and transparency. Future research should therefore assess the legal implications of integrating blockchain technology into a land registration system, specifically on e-DRSA, and examine its compatibility with existing infrastructure to ensure that the socio-technical challenges are addressed for economic benefits to be realised.

This study addresses the theoretical and geographical gap within the existing literature of blockchain technology and a land registration system using an SLR. It contributes to the broader blockchain research in land registration within the South African context and is underpinned by the theoretical foundations. However, the results and implications of this review should be examined taking into consideration its limitations. There are two limitations to this review that are worth noting. Firstly, the scope of the review is limited to the economic benefits of blockchain technology in the land registration system. Consequently, the challenges and risks associated with the applications of blockchain technology in the land registration system fall outside the scope of this review. Secondly, to stay within the defined scope, the literature search was only restricted to land registration and did not encompass land administration functions such as the cadastre system.